An enterprise logistics company runs a continuous 24/7 real-time vehicle tracking pipeline on Google Cloud. The application requires a precise capacity ratio of 3 vCPUs to 24 GB of memory to process telemetry data streams without memory throttling. The workload is strictly non-fault-tolerant and cannot tolerate instance terminations or preemption. Which Compute Engine resource planning strategy provides the most cost-effective configuration while satisfying all performance and operational constraints?
- Provision a Compute Engine VM using a custom machine type configured with 3 vCPUs and 24 GB of RAM, and purchase a Committed Use Discount for the baseline compute resource usage.Cevap
- BProvision a Compute Engine VM using a custom machine type configured with 3 vCPUs and 24 GB of RAM, and enable Spot VM provisioning to reduce hourly runtime costs.
- CProvision a predefined n2-standard-8 instance (8 vCPUs, 32 GB RAM) to meet the memory threshold, relying on automatic Sustained Use Discounts for baseline cost reduction.
- DMigrate the continuous telemetry processing pipeline to Google Cloud Functions with high memory allocation to eliminate virtual machine management overhead.
Cevap
Provision a Compute Engine VM using a custom machine type configured with 3 vCPUs and 24 GB of RAM, combined with a Committed Use Discount for the resource capacity.
Configuring a custom machine type allows precise tailoring of 3 vCPUs and 24 GB of RAM, eliminating unnecessary expenditure on unused vCPUs required by standard machine families. Combining this custom configuration with Committed Use Discounts yields optimal cost reduction for steady-state 24/7 workloads that require guaranteed uptime.
Adım Adım Çözüm
Anahtar Kavram
Planning Compute Engine machine families, custom machine types, and discounting strategies for steady-state workloads.
Tahmini Süre:2m 0s